Donald Trump isn't an economist, but if he were, he'd probably be a German one. (And no, that isn't some kind of Godwin's Law reference).

The simple story is that Trump, like the Germans, thinks that a country's strength comes from the size of its trade surplus — or its weakness from a lack of one. You could hear that in the last debate when he said that "I'm not against trading," but at the same time "I don't want us to lose $58 billion," which is how much more we spend buying things from Mexico than they do from us. It's not all that different from the German Economic Ministry sniffing that their outsized trade surplus is just "a sign of the competitiveness of the German economy and global demand for quality goods from Germany," rather than the result of wrongheaded policies that fetishize selling things to other countries at the expense of investing in their own. In both cases, there's an assumption that economics is some kind of game where the goal is to rack up as many exports as possible. That's Trump's definition of winning.

It's not. And the reason it's not is that we aren't giving Mexico $58 billion for nothing. We're getting oil and trucks and TVs in return. That's why it's called "trade."

To be fair, though, Trump does have a point that you can't talk about trade without talking about exchange rates too. China, for example, really did subsidize its exports by keeping its currency cheaper than it should have been, and that really did cost us jobs. But that's not true anymore. In just the last five years, China has let the yuan rise a lot in real terms and has kept it from falling as much as it would  now that all the money leaving the country is actually putting pressure on it to weaken. Consider this: Beijing spent $107 billion in December alone propping it up. No, the bigger problem now is that the Federal Reserve doesn't seem to be worrying enough about the fact that its plan to raise rates has made the dollar soar some 20 percent in just the last 18 months.

But this might be a little more nuance than Trump wants to entertain. See, it's one thing to say that we shouldn't let other countries cheat or for us to cheat ourselves when it comes to trade. That might mean we need to use diplomatic pressure—not a trade war—to stop countries from exporting unemployment to us, or that the Fed might need to think more about how much a stronger dollar would hurt the economy. But it's another thing to say that the existence of a trade deficit with any one country is in and of itself evidence that something has gone wrong. Or that it somehow means we should get other countries to either pay for a wall or for their military protection or just to enter our market, like Trump did when he proposed slapping a 45 percent tariff on Chinese goods.

That's a mercantilist worldview where the way we win is to force other people to buy our stuff, and the way we lose is to have to buy stuff from them. But that's not the way things actually work. We really are better off with free trade—although I admit that "we" can be a bit of a slippery word. Just like anything else, there are winners and losers from free trade, and for it to be worth it overall, we have to redistribute from the first to the second. But even then we shouldn't kid ourselves: it still isn't easy for communities that lose jobs to find new ones, as economists David Autor, David Dorn, and Gordon Hanson showed was the case with towns that had factory jobs outsourced to China. So there are real costs. But the alternative would be worse. That would be a world where retaliatory tariffs cost us export jobs, things cost more, and it takes more work to make the same amount of stuff.

That's a funny definition of winning.